Defect inspection and charged particle beam apparatus

RELATED APPLICATIONS

This application is a Continuation of U.S. application Ser. No. 11/498,125, filed Aug. 3, 2006, now U.S. Pat. No. 7,348,559 which is a Continuation of U.S. application Ser. No. 11/139,609, filed May 31, 2005, now U.S. Pat. No. 7,112,792, claiming priority of Japanese Application No. 2004-161276, filed May 31, 2004, the entire contents of each of which are hereby incorporated by reference.

BACKGROUND OF THE INVENTION

The present invention relates to a charged particle beam apparatus, for example, an apparatus such as a scanning electron microscope (SEM) for observing a fine pattern on a semiconductor or a general specimen, and a defect inspection apparatus for measuring electrical properties of an electronic device using a fine probe, and more particularly, to a field of view determining method for bringing a probe into contact with a specimen using an image-shift function of a charged particle beam apparatus, and a defect inspection apparatus using the field of view determining method.

Conventionally, known inspection apparatuses for detecting electrical defects in fine electronic circuits formed on semiconductor chips include inspection apparatuses such as an electron beam tester, i.e., EB tester, a probing apparatus, and the like. The EB tester is an apparatus which irradiates an electron beam onto a site under measurement, and detects electrically defective sites of an LSI, taking advantage of the fact that the amount of secondary electrons generated from a site under measurement varies depending on a voltage at the site under measurement. The probing apparatus in turn is an apparatus which brings a plurality of probes or mechanical probes, arranged to match the positions of property measuring pads of an LSI, into contact with measuring pads and plugs to measure the electrical properties of the LSI. With these EB tester and probing apparatus, an operator manually confirms a site with which a probe should be brought into contact, while viewing an image of wires such as an optical microscope (OM) image, a SEM image and the like.

In recent years, increasingly complicated circuit patterns have been formed on semiconductor devices such as LSI's, thereby making it more and more difficult to move a probe to an optimal probing position in a short time. To overcome the difficulties, a technique called “CAD navigation” displays the wiring layout of a semiconductor device in agreement with an actual image of the semiconductor device, referenced by an operator, during a probing operation to reduce a time required for the probing operation.

The SEM image is observed using a scanning electron microscope which scans a primary electron beam on a specimen or semiconductor to capture a scanned image of a fine pattern on the specimen. In order to correctly move a scan area or field of view of the primary electron beam to a point under observation on a specimen, an apparatus intended to observe a fine pattern on the specimen has an image-shift function which electrically deflects the primary electron beam to electrically move a view area in a range of several μm to approximately 10 μm using deflectors systems in series.

Also, since the image-shift function directs the primary electron beam obliquely into an objective lens, off-axis aberrations of the objective lens cause a degradation in the resolution of SEM images. To solve this problem, JP-A-10-247465, for example, discloses a technique for removing the off-axis aberrations as a function of an image-shift amount. Particularly, observations on patterned specimen such as semiconductors are generally made on the order of sub-nanometers or nanometers using low accelerating voltages equal to or lower than 5 kV in order to prevent the specimen from being charged up. When the image-shift function is used for the foregoing purposes under the foregoing conditions, it is necessary to reduce off-axis chromatic aberration and chromatic aberration associated with image-shift deflection.

When the image-shift function is not used, i.e., when the object point of an objective lens does not move, chromatic aberration of the objective lens is effectively reduced to improve the resolution by a retarding method which involves applying a negative voltage to a specimen, or by a boosting method which involves applying a positive voltage into an objective lens. On the other hand, when a specimen is irradiated with an oblique primary electron beam, i.e., in beam tilting, the primary electron beam is intentionally directed out of the axis of an objective lens to generate chromatic aberration, and the chromatic aberration is canceled out using an electrostatic and magnetic multipolor, as disclosed, for example, in JP-A-2001-15055.

While the aforementioned image-shift based movements of the field of view can be substituted by mechanical movements of a DUT stage, the image-shift function is superior in terms of the moving speed and accuracy. Even if the specimen stage (DUT) stage is improved in moving accuracy, mechanical movements cause vibrations at all times. Mechanical vibrations, if any, could damage probes because several probes are often simultaneously brought into contact with measurement plugs during a simultaneous observation in the same SEM field before other probes are brought into contact.

Thus, the operability of a defect analyzer will be significantly improved to reduce a burden on the user if the primary electron beam can be irradiated to a widest possible area, and if the CAD navigation function can be associated with the image-shift function which reduces chromatic aberration.

SUMMARY OF THE INVENTION

In recent years, increasingly complicated circuit patterns have been formed on semiconductor devices such as LSI's, thereby making it more and more difficult to move a probe to an optimal probing position in a short time, and the CAD navigation is effective for quick movements of a probe to an optimal proving position, as has been described above. However, the current CAD navigation only takes into consideration the driving of a DUT stage. The image-shift function is essential to prevent damages of probes due to mechanical vibrations of the DUT stage as mentioned above to accomplish high-speed and accurate movements of the field of view. However, there is no disclosed technique on the CAD navigation linked to the image-shift function.

It is not practically feasible to completely replace a DUT stage driving range by the image-shift. However, a need exists for a widest possible area irradiated with the primary electron beam. As will be described in detail later in connection with embodiments, for example, considering from a sector width of a current semiconductor memory mat, ±75 μm or more is required for an image-shift movement amount, while 150 μm or more is required for a total movement range. Further the size of plugs is required to be equal to or less than 200 nm, and observations should be made at a high resolution of at least 10 k or higher of SEM magnification.

The apparatus disclosed in JP-A-10-247465 can remove off-axis aberration due to a primary electron beam obliquely incident on an objective lens to provide high-resolution SEM images even with an image-shift amount five times as much as the conventional image-shift amount of several μm to approximately 10 μm. However, it is chromatic aberration due to deflection which is prominent with an image-shift of ±75 μm or more, but JP-A-10-247465 does not take into consideration the deflected chromatic aberration.

Even when the image-shift is used, the chromatic aberration can be reduced by the retarding method or boosting method. However, the retarding method involves applying a specimen with a voltage of −1 k volts or higher. In a defect inspection apparatus, probes are directly brought into contact with plugs for electric measurements, so that a voltage applied to a specimen will cause damages of not only the probes and plugs but also an overall device due to a discharge.

The image-shift is similar to the beam tilting, which has been used in recent years for three-dimensional observations on fine patterns of semiconductor devices, in that a primary electron beam is directed obliquely onto a specimen. The apparatus disclosed in JP-2001-15055 intentionally directs a primary electron beam out of the axis of an objective lens to generate chromatic aberration, and also uses an electrostatic and magnetic multipolor to generate chromatic aberration which has the same magnitude as but a different direction from the former chromatic aberration to cancel out the chromatic aberration, so that SEM images can be provided at a high resolution even under a low accelerating voltage condition which involves a large beam inclination. Further, the primary electron beam is controlled to be deflected at all times about an object point on the optical axis of the objective lens in order to minimize the off-axis aberration of the objective lens. However, the field cannot be moved unless the object point is displaced, but a movement of the object point will result in increased off-axis aberration of the objective lens. Thus, the image-shift is essentially different from the beam tilting.

In the present invention, the ratio of an object point of an objective lens to an inclination of a primary electron beam at an image point is defined as the “objective lens axis.” The objective lens axis may be exemplified by a current center axis which is generally known by users of charged particle beam apparatuses. The objective lens axis, as referred to in the present invention, does not exist in the beam tilting since the object point is at center in the beam tilting.

It is therefore an object of the present invention to provide a charged particle beam apparatus which accomplishes an objective lens axis with a reduced degradation in the resolution due to chromatic aberration even with an image-shift of ±75 μm or more, and a defect inspection apparatus which has a CAD navigation function associated with an image-shift function.

The present invention relates to a technique that the coordinate user for using the coordinates for converting an image-shift moving amount to a specimen stage (DUT stage) moving amount in communications between an image processor for processing charged particle beam images and a memory for storing information on circuit line pattern, is introduced into the CAD navigation function, in the defect inspection apparatus which combines probes with a charged particle beam apparatus.

For example, as to a defect inspection apparatus for measuring electric properties of a specimen having a fine circuit line pattern formed on a wafer, wherein a charged particle beam apparatus includes a plurality of probes configured to be brought into contact with a plurality of pads connected to the circuit line pattern or with plugs to measure electric properties of the specimen, means for irradiating the specimen with a charged particle beam, image-shift means for moving a spot irradiated with the charged particle beam on the specimen, means for detecting a secondary charged particle beam generated from the specimen by irradiating the specimen with the charged particle beam to capture an image of the specimen, display means for displaying the image, input means for specifying an arbitrary location on the image, storing means for storing information on the circuit line pattern, image processing means for processing the captured image for displaying the image on the display means, and communication means for interconnecting the storing means and the image processing means, wherein the defect inspection apparatus displays the circuit line pattern and the captured image on the display means, displays information for requesting a user to specify the same location on the circuit line pattern and on the captured image, and communicates information on coordinates of a specified position between the storing means and the image processing means, and the coordinate information includes positional information of the charged particle beam on the specimen by the image-shift means, thereby achieving the aforementioned object.

The present invention significantly improves the user's convenience. Specifically, the present invention can provide a charged particle beam apparatus which reduces a degradation in resolution even with an image-shift of ±75 μm or more, and a defect inspection apparatus which has a CAD navigation function associated with an image-shift function.

Other objects, features and advantages of the invention will become apparent from the following description of the embodiments of the invention taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

<figref idref="DRAWINGS">FIG. 1</figref> is a vertically sectional view illustrating an exemplary configuration of a defect inspection apparatus;

<figref idref="DRAWINGS">FIG. 2</figref> is a schematic diagram illustrating in detail electron optics elements in the defect inspection apparatus illustrated in <figref idref="DRAWINGS">FIG. 1</figref>;

<figref idref="DRAWINGS">FIG. 3</figref> is a schematic diagram illustrating a graphical user interface (GUI) associated with a personal computer (PC) for controlling a SEM of the defect inspection apparatus;

<figref idref="DRAWINGS">FIGS. 4A to 4C</figref> are diagrams each illustrating an exemplary SEM screen on the PC for controlling the SEM of the defect inspection apparatus;

<figref idref="DRAWINGS">FIGS. 5A to 5C</figref> are diagrams each illustrating a GUI for an image-shift unit of the PC for controlling the SEM of the defect inspection apparatus;

<figref idref="DRAWINGS">FIG. 6</figref> shows an exemplary message which is displayed when an image-shift operation amount exceeds a set value;

<figref idref="DRAWINGS">FIGS. 7A and 7B</figref> are flow charts illustrating in combination a basic flow of image-shift control in the defect inspection apparatus illustrated in <figref idref="DRAWINGS">FIG. 1</figref>;

<figref idref="DRAWINGS">FIG. 8</figref> is a flow chart illustrating a basic flow of image-shift control in the defect inspection apparatus illustrated in <figref idref="DRAWINGS">FIG. 1</figref>;

<figref idref="DRAWINGS">FIG. 9</figref> is a conceptual diagram representing the relationship between DUT stage coordinates and image-shift DAC coordinates;

<figref idref="DRAWINGS">FIG. 10</figref> is a conceptual diagram representing the relationship between image-shift operational coordinates and image-shift DAC coordinates; and

<figref idref="DRAWINGS">FIGS. 11A and 11B</figref> are conceptual diagrams for describing a field offset associated with an image-shift axis correction.

DETAILED DESCRIPTION OF THE EMBODIMENTS

In the following, one embodiment of the present invention will be described with reference to the accompanying drawings. <figref idref="DRAWINGS">FIG. 1</figref> illustrates an exemplary configuration of a defect inspection apparatus, and <figref idref="DRAWINGS">FIG. 2</figref> illustrates an exemplary configuration of a scanning electron microscope (hereinafter called the “SEM”). First, the configuration of the defect inspection apparatus will be described with reference to <figref idref="DRAWINGS">FIG. 1</figref>.

SEM electron optics elements, generally designated by <b>101</b>, make up an illumination optical system for irradiating a primary electron beam <b>103</b> onto a specimen and scanning the primary electron beam <b>103</b> on the specimen. Therefore, an electron gun <b>101</b> in this embodiment means a system which includes all components required for the SEM, such as an electron source for generating electron beams, a deflector for scanning a beam, lenses for focusing an electron beam, and the like. A vacuum chamber partition <b>102</b> separates an atmospheric area from a vacuum area. The operation of the SEM electron optics elements <b>101</b>, for example, an electron beam extracting voltage for the electron source, currents applied to the deflector and lenses, and the like are controlled by an electron optics controller <b>116</b>.

Secondary electrons <b>105</b> generated from a specimen <b>118</b> under inspection irradiated with the primary electron beam <b>103</b> are detected by a secondary electron detector <b>104</b>. The secondary electron detector <b>104</b> comprises a sensor unit disposed within the partition <b>102</b> for actually detecting electrons, and a base unit projected out of the partition <b>102</b>, to which wires are connected for connection to a power supply. A mechanical probe <b>106</b> is held by an attachment <b>107</b>, and is brought into contact with a predetermined region of a specimen under inspection. A probe driving means <b>108</b>, for moving the attachment <b>107</b> to a desired position, moves the mechanical probe <b>106</b> together with the attachment <b>107</b> to a desired position.

A specimen, which is actually subjected to a defect inspection, is held on a specimen holder <b>109</b>. The specimen holder <b>109</b> in turn is held by a specimen holder driving means <b>110</b>. The specimen holder <b>109</b> and specimen holder driving means <b>110</b> are collectively called the “DUT stage.”. The DUT stage and probe driving means <b>108</b> are formed on a base stage <b>111</b> which comprises a driving means for integrally driving the DUT stage and probe driving means <b>108</b> in X and Y (in-plane), and Z (vertical) directions. The integral formation of the DUT stage and probe driving means <b>108</b> on the base stage <b>111</b> is one feature of this embodiment. An important aspect, from a viewpoint of technical idea, is to configure the apparatus such that both the specimen <b>118</b> under inspection and mechanical probe <b>106</b> can be moved independently of each other as well as integrally with each other. The base stage <b>111</b> is further carried on a base <b>112</b>.

The specimen holder <b>109</b> and attachment <b>107</b> are connected to an electrical property measuring device <b>113</b>. The electrical property measuring device <b>113</b> mainly measures the current-voltage property of a specimen detected by the mechanical probe <b>106</b>, to calculate a desired property value from the measured property, for example, a resistance, a current value, a voltage, and the like at a location of the specimen in contact with the mechanical probe <b>106</b>. For use in analyses on semiconductor wafers, a semiconductor parameter analyzer may be used, by way of example, for the electrical property measuring device <b>113</b>. The electrical property measuring device <b>113</b> is connected to the specimen base <b>109</b> because a power supply plug may be provided on a specimen carrying surface of the specimen holder <b>109</b> for applying a current or a voltage to the specimen.

The property value measured by the electrical property measuring device <b>113</b> is transmitted to a control computer <b>114</b> through a transmission line. The control computer <b>114</b> makes a higher analysis based on the information transmitted thereto. For example, the control computer <b>114</b> analyzes the measured value to determine whether a measured site is defective or normal. The control computer <b>114</b> is provided with a storing means such as an optical disk drive, a hard disk drive, a memory or the like, so that the measured electrical property value can be stored in the storing means. The control computer <b>114</b> also serves to control the operation of the overall defect inspection apparatus. For example, the electronic gun controller <b>116</b>, secondary electron detector <b>104</b>, probe driving means <b>108</b>, specimen unit, and base stage <b>111</b> operate under the control of the control computer <b>114</b>.

For the purposes mentioned above, the control computer <b>114</b> comprises a memory <b>115</b> for storing software for controlling each of components connected to the control computer <b>114</b>, and an input means for the user to enter set parameters for the defect inspection apparatus. The input means may be, for example, a keyboard, a mouse for moving a pointer on an operation screen, and the like. Data on the wiring layout of a specimen under inspection (hereinafter called the “CAD image data”) is stored in CAD workstation (WS) <b>110</b>. The CAD WS <b>117</b> comprises an image display means for displaying a wiring layout. The CAD WS <b>117</b> is connected to the control computer <b>114</b>, and transmits, as required, CAD image data to the control computer <b>114</b>.

A SEM control personal computer (PC) <b>119</b> controls optical conditions, magnification, focusing, and image-shift for the SEM, the brightness of SEM images, scan speed, alignment, recording of images, movements of the stage and/or probe, and the like in response to operations performed on and commands entered through a graphical user interface (hereinafter called the “GUI”) of the PC or WS. A control panel <b>120</b> implements some of functions of the SEM control PC <b>119</b>, mechanical probe <b>106</b>, specimen unit, and base stage <b>111</b> through operations with knobs, joystick, buttons and the like. It will be apparent that the SEM control PC <b>119</b> may be embodied in a work station.

Referring next to <figref idref="DRAWINGS">FIG. 2</figref>, there is illustrated an exemplary configuration of the SEM electron optics elements. A primary electron beam <b>208</b> emitted and accelerated by an electron gun <b>201</b> is focused by a condenser lens <b>202</b> in front of an aperture <b>203</b>. The amount of the primary electron beam <b>208</b> passing through the aperture <b>203</b> can be adjusted by controlling the position at which the primary electron beam <b>208</b> is focused by the condenser lens <b>202</b>. The primary electron beam <b>208</b>, which has passed through the aperture <b>203</b>, receives a deflecting action applied by image-shift coils <b>204</b>, <b>205</b> and passes through an objective lens <b>209</b>, and is irradiated onto a specimen <b>210</b>. The image-shift coils <b>204</b>, <b>205</b> may be operated independently of each other, or operated with a constant deflection ratio of the upper and lower image-shift coils. In any case, the primary electron beam <b>208</b> can be irradiated to the specimen <b>210</b> at a different position by adjusting the image-shift coils <b>204</b>, <b>205</b>. This operation is generally called a “beam shift” or “image-shift.” Alignment coils <b>206</b>, <b>207</b> are normally adjusted to pass the primary electron beam <b>208</b> through a desired axis of the objective lens <b>209</b>. Also, during an image-shift operation, the alignment coils <b>206</b>, <b>207</b> can adjust an object point <b>211</b> of the objective lens <b>209</b> to correct a field offset during the image-shift operation. This corrective operation will be described later in greater detail.

<figref idref="DRAWINGS">FIG. 2</figref> merely illustrates an exemplary configuration of the SEM electron optics elements. For example, a second condenser lens may be inserted for controlling a convergence angle of the primary electron beam <b>207</b> on the specimen <b>206</b> after it has passed through the aperture <b>203</b>. Also, while the secondary electron detector <b>104</b> is mounted on the vacuum chamber partition <b>102</b> in <figref idref="DRAWINGS">FIG. 1</figref>, a filter and a secondary electron detector may be disposed above the objective lens to extract and detect secondary electrons in a direction opposite to the direction in which the primary electron beam <b>207</b> travels. Further, a boosting electrode may be arranged along the optical axis for improving the resolution.

Referring next to <figref idref="DRAWINGS">FIG. 3</figref>, description will be made on an exemplary graphical user interface (GUI) displayed on the SEM control PC <b>119</b> of the defect inspection apparatus illustrated in <figref idref="DRAWINGS">FIG. 1</figref>. The GUI <b>301</b> on the SEM control PC <b>119</b> is mainly composed of seven windows. A SEM control GUI window <b>302</b> contains icons or a menu of SEM image display, settings of optical conditions for the SEM, magnification of the SEM, focus, image-shift, brightness of SEM images, scan speed, alignment, image recording, and the like. A base stage control GUI window <b>303</b> contains an icon for moving the base stage <b>111</b> to a central position, a CCD observation position, a position at which the mechanical probe <b>106</b> is exchanged, and the like, an icon for removing backlash, and coordinate input/display boxes. A DUT stage control GUI window <b>304</b> contains a cross cursor indicative of a location to which a probe is moved, an arrow-shaped icon, a combo-box for selecting a moving amount, and coordinate input/display boxes. An image-shift control GUI window <b>305</b> contains a cross cursor indicative of a location to which a probe is moved, an arrow-shaped icon, a combo-box for selecting a moving amount, coordinate input/display boxes, and a reset button for returning to an image-shift midpoint, and the like. A coordinate memory display GUI window <b>306</b> contains an icon for registering the base stage coordinates, DUT stage coordinates, and image-shift coordinates in the memory, a combo-box for selecting and calling registered coordinates, and registered coordinate display boxes. A movement selection GUI window <b>307</b> contains an icon for selecting any of the base stage, DUT stage, and shifted image for movement, and a lock icon for preventing any of the base stage, DUT stage, and shifted image from moving. Assume that the foregoing GUI windows are combined into a SEM-stage control GUI complex.

A probe control GUI window <b>308</b> for controlling the driving of the mechanical probe <b>106</b> contains an icon for selecting and displaying a probe unit which the user wishes to drive; an icon for fully retracting a probe; a scroll bar and arrow-shaped icons for driving probes X, Y, Z using a mouse; a scroll bar for finely adjusting a probe driven in the Z-direction with the mouse; reset icons for returning probes X, Y, Z to their respective midpoints; a combo-box for selecting micromotion moving speeds for the probes X, Y, Z; a combo-box for selecting increments for the probes X, Y, Z; a combo-box for selecting a continuous moving speed for the probes X, Y, Z; and a driving state display section for each probe. The probe control GUI window <b>308</b> substantially occupies the right half of the GUI <b>301</b> on the SEM control PC <b>119</b> because a large scroll bar is displayed on the GUI in order to improve the accuracy of operations on the scroll bar through the mouse for driving the probes X, Y, Z.

In <figref idref="DRAWINGS">FIG. 3</figref>, the SEM-stage control GUI complex is displayed on the left side of the GUI, while the probe control GUI window <b>308</b> is displayed on the right side, but the SEM-stage control GUI complex and the probe control GUI window <b>308</b> may be interchanged in place therebetween if such a change facilitates the user's observation and operations.

<figref idref="DRAWINGS">FIGS. 4A</figref>, <b>4</b>B, <b>4</b>C illustrate examples of SEM images displayed on the SEM control GUI window <b>302</b>. In the example illustrated in <figref idref="DRAWINGS">FIG. 4A</figref>, when four probes are used to conduct a defect inspection, four probes are included in the SEM screen field. The probe unit is operated such that a probe <b>401</b> is brought into contact with a target plug <b>405</b> of a plurality of plugs <b>409</b>, and a probe <b>402</b> is brought into contact with a target plug <b>406</b>. A probe <b>403</b> has already been in contact with a target plug <b>407</b>. A probe <b>404</b> is brought into contact with a target plug out of the SEM field <b>410</b>. In the situation illustrated in <figref idref="DRAWINGS">FIG. 4A</figref>, the SEM field <b>410</b> includes the three probes which should be brought into contact with target plugs, in which case the stage need not be moved, or no image-shift operation is required.

In the example illustrated in <figref idref="DRAWINGS">FIG. 4B</figref>, three probes have already been in contact with their respective target plugs. However, a target gate plug <b>408</b> associated with the probe <b>404</b> is out of the SEM field <b>410</b>. In this event, the stage cannot be moved because of mechanical vibrations possibly associated therewith, which could damage the probes in contact. While the stage may be slowly moved in order to prevent damages of the probes, the probes can come off their contacts with the respective target plugs while the stage is being moved, due to a creep phenomenon of piezo devices which are used for driving the probes in order to achieve a positioning resolution of 5 nm or le ss. From the reasons as mentioned above, it is optimal to rely on an image-shift operation to move the SEM field <b>410</b>. An arrow <b>411</b> is displayed to clarify the direction in which one can find the probe <b>404</b> that is selected for driving. The direction in which the arrow is oriented varies depending on the direction in which a probe is driven.

In the example illustrated in <figref idref="DRAWINGS">FIG. 4C</figref>, an image-shift operation has been performed such that the probe <b>404</b> and target gate plug <b>408</b> can be observed within the SEM field <b>410</b>. Current semiconductor devices have plugs, the size of which is 200 nm or less, while a magnification of the SEM for observation is approximately 10 k. With this magnification, an observation field extends approximately to 10 μm. For example, a current semiconductor memory mat has a sector width which is approximately 150 μm at most. If the sector width is made larger to increase the memory capacity, the response degrades due to high frequency components of the device, so that a larger capacity is accomplished by reducing the pitch of a plurality of plugs <b>409</b>. The target gate plug <b>408</b> is located at the end of one side of the sector for pullup, and assuming that the remaining target plugs <b>402</b>-<b>405</b> which one wishes to inspect are located at the opposite end of the sector, the target plugs can be observed with the SEM over the entire range without moving the DUT stage as long as an image-shift moving amount of 150 μm or ±75 μm is satisfied. A conventional SEM may perform an image-shift operation after adjustments of electron optics elements to avoid observing possible contaminations which can stick to the elements during the adjustments of the electron optics elements when the SEM is set to an observation magnification of 50 k or higher for a resolution specimen (like good particles on carbon materials) or the like, and an image-shift moving amount should be approximately 15 μm at most in order to avoid a degradation in the resolution due to off-axis aberration associated with the image-shift operation, and particularly due to deflection chromatic aberration at a low accelerating voltage. Also, for purposes of avoiding observation of contaminations, coordinate positions need not be displayed after the image-shift operation.

However, the defect inspection apparatus lays out a SEM image and a CAD image, so that the defect inspection apparatus is required to have a function for allowing the CAD image to follow a change in the SEM image caused by an image-shift operation, and must have the ability to display an image-shift moving amount in coordinates because there can be a request to bring a probe into contact with a plug located at certain pitches away, even when the CAD image is not used. In the following, this coordinate display capabilities will be described with reference to <figref idref="DRAWINGS">FIGS. 5A</figref>, <b>5</b>B, <b>5</b>C.

<figref idref="DRAWINGS">FIG. 5A</figref> illustrates an example of an image-shift control GUI. DUT stage conversion coordinates <b>501</b> are coordinates of the DUT stage when an image-shift moving amount is converted to a DUT stage moving amount, and are transmitted to the CAD WS <b>117</b> or transmitted from the CAD WS <b>117</b>. A CAD layout image has an area as wide as 10 mm×10 mm, so that even a movement over ±75 μm through an image-shift operation cannot accomplish a movement of the SEM image to fit to the entire CAD layout image through the image-shift operation. Also, a linear scale may be used for the DUT stage to accurately correspond a CAD layout image to the DUT stage coordinates. For this reason, the image-shift coordinates are converted to the DUT stage coordinates which are used for observations. Image-shift operational coordinates <b>502</b> are coordinates for representing an image-shift moving amount which is converted from coordinates at which the image-shift drives (image-shift DAC coordinates) to coordinates at which an operation (observation) is being actually under way, because a raster rotation function is operating for rotating the SEM image field, the image-shift operational coordinates <b>502</b> are required because they are different from the image-shift DAC coordinates. Since the SEM involves an objective lens rotating action, the raster rotation function is operated at all time to control such that a direction in which the DUT stage is moved matches a direction in which the SEM scans. When the user follows the pitch, the user references the image-shift operational coordinates. Details on the coordinate conversion will described later.

An RST <b>503</b> is a button for returning a shifted image to a midpoint. Any of hollow triangular arrow buttons <b>504</b>, when clicked, causes an image to shift in a direction indicated by the clicked arrow button <b>504</b> by a moving amount selected in a combo-box <b>505</b>. The user can enter a desired moving amount into the combo-box <b>505</b>. A cross cursor <b>506</b> permits the user to visually know an actual coordinate position. Image-shift coordinates are displayed in coordinate boxes <b>507</b>, or appropriate values may be directly entered into the coordinate boxes <b>507</b> to shift an image to a desired position. The function described herein is effective to coordinates selected by an option button.

<figref idref="DRAWINGS">FIG. 5B</figref> illustrates another example of the image-shift control GUI. The image-shift control GUI of <figref idref="DRAWINGS">FIG. 5B</figref> additionally comprises a pitch interval entry box <b>508</b>, and counter boxes <b>509</b> for displaying the numbers of pitches in the X- and Y-directions, respectively, in addition to the functions shown in <figref idref="DRAWINGS">FIG. 5A</figref>. When the user enters a pitch into the entry box <b>508</b> and performs an image-shift operation, an integer part of the ratio of an image-shift moving amount to the pitch is displayed in the counter box <b>509</b> for each of the X- and Y-directions. Alternatively, when the user enters the number of pitches by which the user wishes to move an image, and performs an image-shift operation, a message <b>601</b> as shown in <figref idref="DRAWINGS">FIG. 6</figref> is displayed in a message area <b>602</b> when a set number of counts (number of pitches) is reached. A click on a mouse in a Yes area <b>603</b> causes an image-shift operation to continue, whereas a click on a mouse in a No area <b>604</b> results in rejection of the image-shift operation. Since user always performs the image-shift operation for the SEM image field, the counter function is effective only when the image-shift operational coordinates are selected by the option button.

<figref idref="DRAWINGS">FIG. 5C</figref> is an example of the image-shift control GUI which limits the cross cursor, arrow-shaped buttons, combo-box, and counter function only to the image-shift operational coordinates, and displays the specimen conversion coordinates only as reference values. These limitations are placed because the user always operates on the image-shift operational coordinates, except for the entry of coordinates, the DUT stage conversion coordinates are meaningful when a communication is made with the CAD WS <b>117</b>, and the processing is performed within the control computer <b>114</b>.

Next, description will be made on a method of operating the defect inspection apparatus illustrated in <figref idref="DRAWINGS">FIG. 1</figref> using the electron optics components illustrated in <figref idref="DRAWINGS">FIG. 2</figref>. Assume in the following description that positional information has already been calibrated for CAD image data and SEM image data. Details on a coordinate conversion method will be described later.

First, a method of calibrating positional information will be described in brief. The DUT stage is roughly moved to a position corresponding to a point on a certain pattern in a CAD image, and the DUT stage is finely adjusted, while observing on a SEM image, such that the same pattern appears at the center of the screen, to perform an alignment of CAD coordinates to the DUT stage coordinates at a first point. Next, the DUT stage is roughly moved to a position corresponding to a point on another pattern in the same CAD image, and the DUT stage is finely adjusted, while observing on the SEM image, such that the same pattern appears at the center of the screen, to perform an alignment of the CAD coordinates to the DUT stage coordinates at a second point. Finally, the DUT stage is roughly moved to a position corresponding to a point on a further pattern on the same CAD image, and the DUT stage is finely adjusted, while observing on the SEM image, such that the same pattern appears at the center of the screen to perform an alignment of the CAD coordinates to the DUT stage coordinates at a third point. The foregoing operations permit a correspondence to be established between the CAD coordinates and DUT stage coordinates. During the alignment operation, the SEM-based observation may be performed at a low magnification of 100 or lower when the DUT stage is roughly moved, and at a high magnification of approximately 10 k when the DUT stage is finely adjusted. The alignments are preferably performed at three points near corners of the specimen in order to increase the accuracy of the conversion from the CAD coordinates to the DUT stage coordinates over the entire specimen.

Referring first to <figref idref="DRAWINGS">FIGS. 7A</figref>, <b>7</b>B, description will be made on an image-shift operation made by the user through the GUI of the PC <b>119</b>, or with associated knobs and joystick of the control panel <b>120</b>.

As the user performs an image-shift operation through the GUI of the SEM control PC <b>119</b> or with associated knobs and joystick of the control panel <b>120</b> (step <b>701</b>), an operating amount is transmitted to the control computer <b>114</b> which calculates a DAC changing amount for an image-shift XY. The control computer <b>114</b> also calculates a DAC amount of the image-shift XY from a coordinate input amount (step <b>702</b>). For the DAC changing amount, the control computer <b>114</b> adds the DAC changing amount to a current DAC amount or subtract the DAC changing amount from the current DAC amount. Once the DAC amount is determined, the image-shift is actually performed through the electron optics control means <b>116</b> (step <b>703</b>). Also, the control computer <b>114</b> calculates the coordinates on the image-shift DAC axes from an actually set image-shift DAC value (step <b>704</b>).

Subsequently, two different sequence of steps may be contemplated. A first sequence of steps converts the image-shift DAC coordinates to the DUT stage coordinates, and a second sequence of steps converts the image-shift DAC coordinates to the image-shift operation axes. First described is the sequence of steps for converting the image-shift DAC coordinates to the DUT stage coordinates.

The control computer <b>114</b> converts the image-shift DAC coordinates to the DUT stage coordinates (step <b>705</b>). The converted coordinate values are transmitted to the SEM control PC <b>119</b> for displaying the coordinates on the GUI (step <b>706</b>). In this event, when the CAD WS <b>117</b> is not linked to the SEM control PC <b>119</b> so that a CAD layout image is not overlaid on a SEM image (step <b>707</b>), it is determined again whether or not an image-shift operation occurs (step <b>716</b>).

Conversely, when the CAD WS <b>117</b> is linked to the SEM control PC <b>119</b> so that the CAD layout image is overlaid on the SEM image (step <b>707</b>), the control computer <b>114</b> converts the DUT stage coordinates to CAD coordinates and transmits a moving amount to the CAD WS <b>117</b> (step <b>708</b>) before it receives a new CAD image from the CAD WS <b>117</b> (step <b>709</b>), and overlays the CAD image on the GUI of the SEM control PC <b>119</b> (step <b>710</b>). Subsequently, the flow proceeds to step <b>707</b>. It is determined again whether or not an image-shift operation occurs (step <b>716</b>), and the processing is terminated when no image-shift operation occurs (step <b>716</b>), or the flow returns to step <b>702</b> when the image-shift operation occurs (step <b>716</b>), such that the control computer <b>114</b> again performs similar processing to the foregoing.

Next described is the sequence of steps for converting the image-shift DAC coordinates to the image-shift operation axes. The control computer <b>114</b> converts the image-shift DAC coordinates to the image-shift operational coordinates (step <b>711</b>). Then, the control computer <b>114</b> transmits the converted coordinate values to the SEM control PC <b>119</b> for displaying the coordinates on the GUI (step <b>711</b>).

The control computer <b>114</b> determines whether or not the counters have been set on the image-shift operational coordinate axes on the SEM control PC <b>119</b> for displaying the number of pattern widths by which a movement has been made in the X- and Y-directions, respectively. When the counters are not set (step <b>713</b>), the flow proceeds to step <b>707</b>. Conversely, when the counters have been set (step <b>713</b>), the control computer <b>114</b> calculates the ratio of each of the calculated image-shift operational coordinates to the pattern width selected on the GUI of the SEM control PC <b>119</b>, removes the decimal point from the calculated ratio, and defines the resultant value as a count value in the X- or Y-direction (step <b>714</b>). The control computer <b>114</b> transmits the count values to the SEM control PC <b>119</b> for display on the GUI (step <b>715</b>). Subsequently, the flow proceeds to step <b>707</b>.

Also, the SEM control PC <b>119</b> sets the number of pattern widths by which a movement is made in each of the X- and Y-directions, or the SEM control PC <b>119</b> sets an image-shift amount by which a movement is made on each of the image-shift operational coordinate axes, and the control computer <b>114</b> compares actual count values with the set values. When the set values are exceeded by the actual count values, the image of <figref idref="DRAWINGS">FIG. 6</figref> is displayed for permitting the user to determine whether or not the image-shift operation is continued.

Referring next to <figref idref="DRAWINGS">FIG. 8</figref>, description will be made on the image-shift operation when the user moves a layout pattern on the CAD WS <b>117</b>.

A moving amount by which a layout pattern has been moved on the CAD WS <b>117</b> is transmitted to the control computer <b>114</b> (step <b>801</b>). The control computer <b>114</b> calculates a moving amount on the image-shift coordinates, as converted to the DUT stage axes (step <b>802</b>). When the specimen conversion coordinates are directly inputted on the GUI of the SEM control PC <b>119</b>, the control is started from this moment (step <b>803</b>). Next, the moving amount is transmitted to the control computer <b>114</b> which calculates a moving amount on the image-shift DAC axes. The control computer <b>114</b> also calculates input coordinate values on the image-shift DAC axes from the input coordinate amounts (step <b>804</b>). Next, the control computer <b>114</b> calculates a DAC change amount from the moving amount, and adds the calculated DAC change amount to a current DAC amount or subtracts the DAC change amount from the current DAC amount. The control computer <b>114</b> calculates the DAC amount from the input coordinate values. Once the DAC amount is determined, an image-shift is actually performed through the electron optics control means <b>116</b> (step <b>805</b>). Also, the control computer <b>114</b> again calculates the coordinates on the image-shift DAC axes from the actually set image-shift DAC value (step <b>806</b>).

The control computer <b>114</b> converts the image-shift DAC coordinates to the DUT stage coordinates (step <b>807</b>). The converted coordinate values are transmitted to the SEM control PC <b>119</b> for displaying the coordinates on the GUI (step <b>808</b>). In this event, when the CAD WS <b>117</b> is not linked to the SEM control PC <b>119</b>, so that the CAD layout image is not overlaid on the SEM image (step <b>809</b>), the processing is terminated (step <b>813</b>).

Conversely, when the CAD WS <b>117</b> is linked to the SEM control PC <b>119</b>, so that the CAD layout image is overlaid on the SEM image (step <b>809</b>), the control computer <b>114</b> converts the DUT stage coordinates to the CAD coordinates and transmits a moving amount to the CAD WS <b>117</b> (step <b>810</b>) before it receives a new CAD image from the CAD WS <b>117</b> (step <b>811</b>), and overlays the CAD image on the GUI of the SEM control PC <b>119</b> (step <b>812</b>), followed by termination of the processing (step <b>813</b>).

By executing the foregoing steps, a CAD image and coordinates overlaid on the GUI of the SEM control PC <b>119</b> can be automatically updated following a change in the SEM image field resulting from an image-shift operation, thus significantly reducing a burden on the user during the probing.

Next, description will be made on the coordinate conversion associated with the image-shift described in the foregoing embodiment. The following six sets of coordinates should be taken into consideration in the image-shift control:

(1) DUT stage coordinates (x_DUT, y_DUT);

(2) Image-shift coordinates converted to DUT stage (x_IS-DUT, y_IS-DUT);

(3) Image-shift axis correction coordinates (X_IS, Y_IS);

(4) Ideal image-shift coordinates (x_IS, y_IS);

(5) Image-shift DAC coordinates (x_IS-DAC, y_IS-DAC); and

(6) Image-shift operational coordinates (x_IS-OP, y_IS-OP)

The DUT stage coordinates (1) and image-shift coordinates converted to the DUT stage (2) are basically completely the same coordinates if the linearity and orthogonality can be ignored for the DUT stage. The image-shift axis correction coordinates (3) are coordinates converted when an image-shift coil is ideally disposed. The ideal image-shift coordinates (4) are similar to the image-shift axis correction coordinates (3), and are coordinates which are converted in consideration of a field offset which occurs when an off-axis due to an image-shift is corrected. The image-shift DAC coordinates (5) are coordinates converted when the orthogonality of the image-shift coil deviates from an ideal axis in the ideal image-shift coordinates (4). The image-shift operational coordinates (6) are coordinates converted to fit the coordinate axes of the image-shift axis correction coordinates (3) to the SEM scan axis, and are used for correcting the rotation of the scan axis caused by excitation of the objective lens to the coordinate axes of the DUT stage, and for electrically rotating the scan axis.

As described in connection with <figref idref="DRAWINGS">FIGS. 7A</figref>, <b>7</b>B, <b>8</b>, conversion in coordinates required for controlling an image-shift operation involve the relationships of a conversion between the coordinates (x_IS-DUT, y_IS-DUT) (2) and the coordinates (x_IS-DAC, y_IS-DAC) (5) and a conversion between the coordinates (x_IS-OP, y_IS-OP) (6) and the coordinates (x_IS-DAC, y_IS-DAC) (5). The respective relationships of the coordinate conversions are described in <figref idref="DRAWINGS">FIGS. 9 and 10</figref>, respectively. The respective coordinate conversions can be expressed in mathematical formulae as follows:

<maths id="MATH-US-00001" num="00001"><math overflow="scroll"><mtable><mtr><mtd><mtable><mtr><mtd><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mtd><mtd><mo>=</mo></mtd><mtd><mrow><mi>A</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>X</mi><mi>IS</mi></msub></mtd></mtr><mtr><mtd><msub><mi>Y</mi><mi>IS</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>X</mi><mi>IS</mi></msub></mtd></mtr><mtr><mtd><msub><mi>Y</mi><mi>IS</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow></mtd><mtd><mo>=</mo></mtd><mtd><mrow><mi>B</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>IS</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>IS</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>IS</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>IS</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow></mtd><mtd><mo>=</mo></mtd><mtd><mrow><mi>C</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd></mtr><mtr><mtd><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mtd><mtd><mo>=</mo></mtd><mtd><mrow><mi>D</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>X</mi><mi>IS</mi></msub></mtd></mtr><mtr><mtd><msub><mi>Y</mi><mi>IS</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mtd></mtr></mtable></mtd><mtd><mrow><mi>Equations</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>1</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0001.tif" />

A matrix A is a matrix of linear transformation of the image-shift coordinates converted to the DUT stage and the image-shift axis correction coordinates, and depends on a rotating angle due to excitation of the objective lens. A matrix B is a matrix of linear transformation of the image-shift axis correction coordinates and ideal image-shift coordinates, and is a matrix which takes into consideration a field offset which occurs when an off-axis during an image-shift operation is corrected by a single alignment coil or image-shift coil. A matrix C is a matrix of linear transformation of the ideal image-shift coordinates and image-shift DAC coordinates, and depends on an orthogonality angle of the image-shift coil. A matrix D is a matrix of linear transformation of the image-shift operational coordinates and image-shift axis correction coordinates, and depends on an electric rotating angle of the scan axis due to raster rotation in addition to a rotating angle due to excitation of the objective lens. When the image-shift axis can also be rotated, the matrix D depends on a mechanical offset angle of the scan axis from the ideal image-shift axis.

From the foregoing relationships, transformations can be derived for (2) and (5) and for (6) and (5).

<maths id="MATH-US-00002" num="00002"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><mi>ABC</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow><mo>⁢</mo><mstyle><mtext></mtext></mstyle><mo>⁢</mo><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><msup><mrow><mo>(</mo><mi>ABC</mi><mo>)</mo></mrow><mrow><mo>-</mo><mn>1</mn></mrow></msup><mo>⁢</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equations</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>2</mn></mrow></mtd></mtr><mtr><mtd><mrow><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><mi>DBC</mi><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow><mo>⁢</mo><mstyle><mtext></mtext></mstyle><mo>⁢</mo><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DAC</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><msup><mrow><mo>(</mo><mi>DBC</mi><mo>)</mo></mrow><mrow><mo>-</mo><mn>1</mn></mrow></msup><mo>⁢</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>OP</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equations</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>3</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0002.tif" />

Some aspects to be noted for each of matrix elements will be described below.

The actual DUT stage coordinates and the image-shift coordinates converted to the DUT stage may suffer from a zero offset depending on the linearity and orthogonality of the DUT stage and the positioning of the secondary electron detector of the SEM, and can therefore fail to fit to each other. In such a case, the zero offset can be corrected for by the following relationship:

<maths id="MATH-US-00003" num="00003"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>DUT</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>DUT</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><mrow><msup><mi>A</mi><mi>′</mi></msup><mo>⁡</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow><mo>+</mo><mrow><mo>(</mo><mtable><mtr><mtd><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mrow></mtd></mtr><mtr><mtd><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mrow></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>4</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0003.tif" />

A matrix A′ is a matrix of linear transformation which takes into consideration the linearity and orthogonality of the DUT stage. Δx_IS-DUT, Δy_IS-DUTrepresent offset amounts of the DUT stage from a spot irradiated with a primary electron beam. In an actual apparatus, a relative movement from fixed conditions is often relied on to handle the offset amounts. Also, in an image-shift operation in a range of approximately ±75 μm as compared with a wide operation range of the DUT stage extending 10 mm or more, the linearity and orthogonality of the DUT stage can be often ignored. Therefore, the matrices A and A′ are treated as the same coordinates in the following description.

The matrices B, C comprise elements which are not prominent in a normal image-shift range of ±15 μm, as described in connection with <figref idref="DRAWINGS">FIGS. 4A</figref>, <b>4</b>B, <b>4</b>C, but are prominent when an image-shift range of ±75 μm or more is accomplished, as required by the defect inspection apparatus, and which must be taken into consideration for highly accurate positional control. The matrix B can be equal with a unit matrix when a correction is added to prevent the object point of the objective lens from being moved by two alignment coils or an image-shift coil. In the following, the matrix B will be described with reference to <figref idref="DRAWINGS">FIGS. 11A</figref>, <b>11</b>B.

<figref idref="DRAWINGS">FIG. 11A</figref> conceptually illustrates a ray trajectory during the image-shift control. An actual trajectory of a primary electron beam is represented by a ray trajectory <b>1</b>. When image-shift coils <b>204</b>, <b>205</b> are driven, the ray trajectory <b>1</b> changes to a virtual ray trajectory <b>1</b>. When an upper alignment coil <b>206</b> alone is further driven, the ray trajectory <b>1</b> changes to a combination of a virtual ray trajectory <b>2</b> and a ray trajectory <b>2</b>. When a lower alignment coil <b>207</b> is further driven, the ray trajectory <b>1</b> changes to a combination of a virtual ray trajectory <b>3</b> and a beam trajectory <b>3</b>.

<figref idref="DRAWINGS">FIG. 11B</figref> only illustrates central trajectories of the ray trajectories shown in <figref idref="DRAWINGS">FIG. 11A</figref>. The primary electron beam, emitted from a primary electron beam crossover point (object point A<b>1</b> of the objective lens) on the surface of the objective lens, is deflected by the upper image-shift coil <b>204</b> (B<b>1</b>), is again deflected by the lower image-shift coil <b>205</b> in the opposite direction (C<b>1</b>), passes through one point (D<b>1</b>) on the main surface of the objective lens, and is irradiated obliquely to an objective lens image point (E<b>1</b>) on an objective lens image plane (i.e., the specimen). As the image-shift coils are driven, the object point A<b>1</b> of the objective lens is shifted to A<b>2</b>, so that the trajectory virtually follows A<b>2</b>-B<b>2</b>-C<b>1</b>-D<b>1</b>-E<b>1</b>. When the upper alignment coil <b>206</b> alone is driven in this state, the object point A<b>2</b> of the objective lens is shifted to A<b>3</b>, so that the trajectory changes to A<b>3</b>-B<b>2</b>-C<b>2</b>-D<b>2</b>-E<b>2</b>. The shift of the object point A<b>2</b> of the objective lens to A<b>3</b> causes a shift of the objective lens image point E<b>1</b> to E<b>2</b>, so that the primary electron beam is irradiated to an offset spot. Further, when the lower alignment coil <b>207</b> is also driven in this state to shift the object point A<b>3</b> of the objective lens to fit to A<b>3</b>, the resulting trajectory follows A<b>2</b>-B<b>2</b>-C<b>2</b>-D<b>3</b>-E<b>1</b>. When the object point A<b>3</b> of the objective lens is fitted to A<b>2</b>, the primary electron beam is irradiated to the original objective lens image point E<b>1</b>, so that the offset of the irradiated spot can be removed.

The image-shift in SEM is basically relied on a method of moving a spot on a specimen irradiated with a primary electron beam by deflecting the primary electron beam using two image-shift coils. When the primary electron beam is deflected by an image-shift coil, the primary electron beam obliquely impinges on a specimen, resulting in off-axis aberration. Also, chromatic aberration is produced in association with the deflection. For minimizing the influence of such aberrations, the image-shift control is conducted such that a desired axis is found by adjusting the ratio of currents applied to the upper and lower image-shift coils, or currents independently applied to the respective coils in the X-direction and Y-direction, respectively. Actually, however, the desired axis cannot be found only with adjustments in the same direction by a rotating action of the objective lens. For this reason, a single alignment coil may be provided independently of the image-shift coils, or an alignment signal is superimposed on the upper image-shift coil to add a correction depending on an image-shift amount, thereby achieving the identification of the desired axis.

As described above, a correction made by a single alignment coil causes a shift of the object point of the objective lens and a resulting shift of a spot on a specimen irradiated with the primary electron beam to give rise to a field displacement, thus resulting in coordinates which are different from those to which one wishes to actually shift the primary electron beam. In consideration of this amount, the matrix B is set in order to shift the primary electron beam to desired image-shift coordinates. However, it is possible to avoid the primary electron beam shifted from an intended irradiated spot on the specimen and remove the field movement by adding a correction depending on an image-shift amount to an alignment signal by providing two independent alignment coils or superimposing the alignment signal on the upper and lower image-shift coils to find the desired axis, and simultaneously conducting the image-shift control to prevent the object point of the objective lens from shifting. When the field movement can be removed by this method, the matrix A can be treated as a unit matrix, without the need for taking into consideration the matrix B, resulting in an improvement in the coordinate conversion accuracy.

The desired axis, herein referred to, may be exemplified by a current center axis. The current center axis refers to an axis on which a change in excitation of an objective lens does not cause a change in the position of the image point of the objective lens. Among the off-axis aberrations, astigmatism can be corrected by an astigmatism correcting coil, not shown in <figref idref="DRAWINGS">FIG. 2</figref>, while field curvature aberration can be corrected by adjusting the excitation of the objective lens. Though depending on the characteristics of a particular objective lens and electron optics conditions, an image-shift of 100 μm will cause coma aberration of approximately 2-3 nm to leave on the current center axis. However, it is chromatic aberration resulting from the image-shift, which exerts larger influences. Specifically, the chromatic aberration is prominent particularly at low accelerating voltages equal to or lower than 5 kV. For example, in a charged particle beam apparatus which is equipped with a field emission electron source having an energy spread of 0.3 eV, an image-shift of 100 μm at an accelerating voltage of 2 kV causes chromatic aberration of approximately 14 nm. Particularly, in the defect inspection apparatus which is intended for semiconductor materials as specimen, observations at low accelerating voltages are essential in order to minimize damages to specimen. Thus, an axis, called the “achromatic axis,” which can remove off-axis chromatic aberration and deflection chromatic aberration of an objective lens, can be assumed as a desired axis. The aberrations given herein as examples differ from one charged particle beam apparatus to another, as will be apparent to users of charged particle beam apparatuses.

For conditions to find a desired axis, the current center axis is generally well known in charged particle beam apparatuses. In the following, the achromatic axis will be described in detail with reference to appropriate equations. <br /><i>w</i>₀<i>=a·w</i>₀′  Equation 5<br /> where wo, wo′ are the position and inclination of the primary electron beam on the surface of the objective lens, and a represents conditions for a desired axis. The position wi and inclination wi′ of the primary electron beam on the image plane of the objective lens can be expressed by the following equations using Equation 5: <br /><i>w</i>_i<i>=a·g</i>_i<i>·w</i>₀′<br /><i>w</i>_i′=(<i>a·g</i>_i<i>′+h</i>_i′)·<i>w</i>₀′  Equations 6<br /> where gi, hi, gi′, hi′ are the positions and inclinations at the image points of two paraxial rays well known in the charged particle theory. In the charged particle theory, chromatic aberration of a primary electron beam obliquely incident on an objective lens is expressed by the following equation:

<maths id="MATH-US-00004" num="00004"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><msub><mi>w</mi><mi>c</mi></msub></mrow><mo>=</mo><mrow><mrow><mo>{</mo><mrow><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mn>0</mn></mrow></msub><mo>·</mo><msubsup><mi>w</mi><mi>i</mi><mi>′</mi></msubsup></mrow><mo>+</mo><mrow><mrow><mo>(</mo><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>m</mi></mrow></msub><mo>+</mo><msub><mi>jC</mi><mi>cr</mi></msub></mrow><mo>)</mo></mrow><mo>·</mo><msub><mi>w</mi><mi>i</mi></msub></mrow></mrow><mo>}</mo></mrow><mo>·</mo><mrow><mo>(</mo><mfrac><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>V</mi></mrow><mi>V</mi></mfrac><mo>)</mo></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>7</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0004.tif" /><br /> where Cco is an axial chromatic aberration coefficient, Ccm is a magnification chromatic aberration coefficient, Ccr is an anisotropic chromatic aberration coefficient, V is the energy of the primary electron beam on the objective lens image plane, and ΔV is variations in the energy of the primary electron beam. An image-shift causes variations ΔV in the energy which result in variations in the position to which the primary electron beam propagates. The variations can be regarded as a type of aberration, which is called the “deflection chromatic aberration.” This deflection chromatic aberration is not included in Equation 7. For example, in a charged particle beam apparatus equipped with a field emission electron source having an energy width of 0.3 eV, an image-shift of 100 μm at an accelerating voltage of 2 kV causes deflection chromatic aberration of approximately 8 nm, as compared with the chromatic aberration expressed by Equation 7 which is approximately 6 nm. While it is possible to find from Equation 7 axial conditions under which the chromatic aberration disappears, deflection chromatic aberration of 8 nm remains in the foregoing example. It is therefore necessary to find an axis a which satisfies conditions under which the chromatic aberration is eliminated, in consideration of the deflection chromatic aberration as well. The chromatic aberration produced by a deflecting action such as the image-shift can be expressed by the following equation:

<maths id="MATH-US-00005" num="00005"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><msub><mi>w</mi><mi>def</mi></msub></mrow><mo>=</mo><mrow><mrow><msub><mi>C</mi><mi>def</mi></msub><mo>·</mo><msub><mi>r</mi><mi>IS</mi></msub><mo>·</mo><mrow><mo>(</mo><mfrac><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>V</mi></mrow><mi>V</mi></mfrac><mo>)</mo></mrow></mrow><mo>=</mo><mrow><msub><mi>C</mi><mi>def</mi></msub><mo>·</mo><mi>a</mi><mo>·</mo><msub><mi>g</mi><mi>i</mi></msub><mo>·</mo><msubsup><mi>w</mi><mn>0</mn><mi>′</mi></msubsup><mo>·</mo><mrow><mo>(</mo><mfrac><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>V</mi></mrow><mi>V</mi></mfrac><mo>)</mo></mrow></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>8</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0005.tif" /><br /> where:

<maths id="MATH-US-00006" num="00006"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mrow><msub><mi>C</mi><mi>def</mi></msub><mo>⁡</mo><mrow><mo>(</mo><mrow><mi>MAGNETIC</mi><mo>⁢</mo><mstyle><mspace width="0.8em" height="0.8ex" /></mstyle><mo>⁢</mo><mi>FIELD</mi></mrow><mo>)</mo></mrow></mrow><mo>=</mo><mrow><mrow><mo>-</mo><mfrac><mn>1</mn><mn>2</mn></mfrac></mrow><mo>·</mo><mrow><mo>(</mo><mfrac><mi>V</mi><msub><mi>V</mi><mi>def</mi></msub></mfrac><mo>)</mo></mrow></mrow></mrow><mo>⁢</mo><mstyle><mtext></mtext></mstyle><mo>⁢</mo><mrow><mrow><msub><mi>C</mi><mi>def</mi></msub><mo>⁡</mo><mrow><mo>(</mo><mrow><mi>ELECTRO</mi><mo>⁢</mo><mstyle><mtext>-</mtext></mstyle><mo>⁢</mo><mi>STATIC</mi><mo>⁢</mo><mstyle><mspace width="0.8em" height="0.8ex" /></mstyle><mo>⁢</mo><mi>FIELD</mi></mrow><mo>)</mo></mrow></mrow><mo>=</mo><mrow><mo>-</mo><mrow><mo>(</mo><mfrac><mi>V</mi><msub><mi>V</mi><mi>def</mi></msub></mfrac><mo>)</mo></mrow></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>9</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0006.tif" /><br /> where C_defis a chromatic aberration coefficient due to deflection, and takes different values for magnetic deflection and electro-static deflection, ris is an image-shift operation amount, and Vdef is the energy of the primary electron beam at a deflected location. Combining Equation 9 with Equation 10 results in the following equation:

<maths id="MATH-US-00007" num="00007"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><msub><mi>w</mi><mi>c</mi></msub></mrow><mo>=</mo><mrow><mrow><mo>{</mo><mrow><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mn>0</mn></mrow></msub><mo>·</mo><mrow><mo>(</mo><mrow><mrow><mi>a</mi><mo>·</mo><msubsup><mi>g</mi><mi>i</mi><mi>′</mi></msubsup></mrow><mo>+</mo><msubsup><mi>h</mi><mi>i</mi><mi>′</mi></msubsup></mrow><mo>)</mo></mrow></mrow><mo>+</mo><mrow><mrow><mo>(</mo><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>m</mi></mrow></msub><mo>+</mo><msub><mi>jC</mi><mi>cr</mi></msub></mrow><mo>)</mo></mrow><mo>·</mo><mi>a</mi><mo>·</mo><msub><mi>g</mi><mi>i</mi></msub></mrow><mo>+</mo><mrow><msub><mi>C</mi><mi>def</mi></msub><mo>·</mo><mi>a</mi><mo>·</mo><msub><mi>g</mi><mi>i</mi></msub></mrow></mrow><mo>}</mo></mrow><mo>·</mo><msubsup><mi>w</mi><mn>0</mn><mi>′</mi></msubsup><mo>·</mo><mrow><mo>(</mo><mfrac><mrow><mi>Δ</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>V</mi></mrow><mi>V</mi></mfrac><mo>)</mo></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>10</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0007.tif" /><br /> The axis a which satisfies the condition for deriving zero from Equation 10 is the achromatic axis which can be expressed by the following equation:

<maths id="MATH-US-00008" num="00008"><math overflow="scroll"><mtable><mtr><mtd><mrow><mi>a</mi><mo>=</mo><mrow><mo>-</mo><mfrac><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mn>0</mn></mrow></msub><mo>·</mo><msubsup><mi>h</mi><mi>i</mi><mi>′</mi></msubsup><mo>·</mo></mrow><mrow><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mn>0</mn></mrow></msub><mo>·</mo><msubsup><mi>g</mi><mi>i</mi><mi>′</mi></msubsup></mrow><mo>+</mo><mrow><mrow><mo>(</mo><mrow><msub><mi>C</mi><mrow><mi>c</mi><mo>⁢</mo><mstyle><mspace width="0.3em" height="0.3ex" /></mstyle><mo>⁢</mo><mi>m</mi></mrow></msub><mo>+</mo><msub><mi>C</mi><mi>def</mi></msub></mrow><mo>)</mo></mrow><mo>·</mo><msub><mi>g</mi><mi>i</mi></msub></mrow><mo>+</mo><mrow><msub><mi>jC</mi><mi>cr</mi></msub><mo>·</mo><msub><mi>g</mi><mi>i</mi></msub></mrow></mrow></mfrac></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>11</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0008.tif" />

Among the equations described above, wo, wo′, wi, wi′, a, rIS take complex numbers, while the remainings take real numbers. Also, the exemplary values given above differ depending on the characteristics of the objective lens.

As previously described, a correction depending on an image-shift amount is added to an alignment signal to create an achromatic axis by providing two independent alignment coils or superimposing an alignment signal on the upper and lower image-shift coils, and simultaneously, control is conducted to prevent the object point of the objective lens from shifting, thereby making it possible to avoid the primary electron beam irradiated to a shifted position on a specimen and effectively eliminate the chromatic aberration.

Published Japanese Translation of PCT International Publication for Patent Application WO 01/033603 describes an ExB field generator, i.e., an apparatus which generates an energy distribution, which has the same magnitude as and a direction reverse to off-axis/deflection chromatic aberration corresponding to an image-shift amount, using a wien filter to cancel out aberration. The use of this strategy will result in the realization of a current center axis of the objective lens to reduce off-axis aberration such as coma aberration and an additional reduction in off-axis/deflection chromatic aberrations. Presumably, the use of the ExB field generator may involve a complicated configuration and control of the apparatus and a high cost.

Finally, description will be made on coordinate conversion used in communications between the SEM control PC <b>119</b> and CAD WS <b>117</b>. Coordinates which should be taken into consideration in the CAD navigation include the following four sets:

(1) DUT stage coordinates (x_DUT, y_DUT);

(2) Base stage coordinates (x_BASE, y_BBASE);

(3) Image-shift coordinates converted to DUT stage (x_IS-DUT, y_IS-DUT); and

(4) CAD navigation coordinates (x_CAD, y_CAD)

The CAD navigation coordinates (4) define layout coordinates for a CAD image which has an ideal magnification and pattern information. The DUT stage coordinates (1) represent a linear scale value of the DUT stage. However, the coordinates also include an correction term for small errors due to backlash of stage mechanisms and the like. The base stage coordinates (3) are coordinates resulting from a conversion from an image-shift amount to the DUT stage axis. The coordinate conversion is performed according to the following equation:

<maths id="MATH-US-00009" num="00009"><math overflow="scroll"><mtable><mtr><mtd><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>CAD</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>CAD</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>=</mo><mrow><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>DUT</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>DUT</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>+</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mi>BASE</mi></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mi>BASE</mi></msub></mtd></mtr></mtable><mo>)</mo></mrow><mo>+</mo><mrow><mo>(</mo><mtable><mtr><mtd><msub><mi>x</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr><mtr><mtd><msub><mi>y</mi><mrow><mi>IS</mi><mo>-</mo><mi>DUT</mi></mrow></msub></mtd></mtr></mtable><mo>)</mo></mrow></mrow></mrow></mtd><mtd><mrow><mi>Equation</mi><mo>⁢</mo><mstyle><mspace width="1.1em" height="1.1ex" /></mstyle><mo>⁢</mo><mn>12</mn></mrow></mtd></mtr></mtable></math></maths><img file="US7705303B2_D0009.tif" />

When the DUT stage and base stage imply rotational misalignment and zero offset, a conversion correction may be made in a manner similar to Equation 4.

While the foregoing embodiment has been described in connection with a scanning electron microscope which is given as an example of a charged particle beam apparatus, it will be apparent to developers and users of charged particle beam apparatuses that the same description can be applied to a focused ion beam apparatus for observing and processing specimen using ion beams. Also, due to the use of heavy elements such as Ga ions, an electrostatic lens is used for the objective lens instead of a magnetic lens. Thus, the matrix A in Equation 1 need not be taken into consideration because there is no rotating action, as experienced with a magnetic objective lens. Also, since the rotating action of the objective lens can also be ignored in the matrix D as is the case with the matrix A, it can be readily contemplated that the image-shift control accuracy is improved.

The present invention proposes a charged particle beam apparatus which can control an image-shift of ±75 μm or more without a degraded resolution due to chromatic aberration, and can display and input coordinates by a conversion from the stage coordinates to the image-shift control coordinates, a conversion from the image-shift operational coordinates to the image-shift control coordinates, and the realization of the achromatic axis. The present invention also proposes an image-shift control method suitable for the introduction of CAD navigation into a defect inspection apparatus which is a combination of a probe with the charged particle beam apparatus. According to the present invention, the user's convenience is remarkably improved when the user uses an image-shift function of the charged particle beam apparatus and defect inspection apparatus.

It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.